Analysis

Visualization

Visualization tools are an important part of the exploratory modeling process. They allow analysts and other stakeholders to conceptualize the potentially complex relationships represented in transportation models.

• Scatter Plot Matrix
  • Contrasting Sets of Experiments
  • Scatter Plot Matrix API
• Identifying Model Problems
• Interactive Visualizer
  • Single Dimension Figures
  • Changing Selections
  • Two Dimension Figures
  • Dynamic Feature Scoring
  • Interactive PRIM
  • Visualizer API
    • Manipulating Selections
    • Single Dimension Views
    • Multi Dimension Views
    • Analytical Tools

Scenario Discovery

Scenario discovery in exploratory modeling is focused on finding scenarios that are interesting to the user. The process generally begins through the identification of particular outcomes that are “of interest”, and then discovering what factor or combination of factors can result in those outcomes. There are a variety of methods to use for scenario discovery. We illustrate a few here.

• Feature Scoring
  • Road Test Feature Scores
  • Interpreting Feature Scores
  • Threshold Scoring
  • Feature Scoring API
• PRIM
  • The Mechanics of using PRIM
  • Non-Rectangular Boxes
  • PRIM API
• CART
  • The Mechanics of using CART
  • CART API

Directed Search

The scenario discovery tools outlined above are focused on exploring parameters and outcomes across a pre-set sample of model runs. This sample can be small or quite large, but the tools only consider cases that have already been evaluated. In directed search, we transition to tools that will propose and execute new cases.

• Policy Contrast
• Optimization Tools
  • Multi-Objective Optimization
  • Robust Optimization
    • Robustness Functions
• Mechanics of Using Optimization
  • Policy Optimization: Search over Levers
  • Worst Case Discovery: Search over Uncertainties
  • Using Robust Optimization
    • Constraints